Autism is a severe developmental disorder characterized by deficits in social interaction and communication and by stereotypic or repetitive behaviors or restricted interests. Despite a strong influence of genetics, environmental factors are likely to play a causal role for many children. The disorder is defined purely by behavioral phenotype, with research only now burgeoning on biologic mechanisms and biochemical markers of pathogenesis. The proposed project will continue an epidemiologic study of the environmental and genetic causes of autism that was initiated under the UC Davis Center for Children's Environmental Health (CCEH) funded in the fall of 2001 by the NIEHS (1 P01 ES11269), U.S. EPA, and M.I.N.D. (Medical Investigations of Neurodevelopmental Disorders) Institute. This project, known as the CHARGE (Childhood Autism Risks from Genetics and Environment) Study, is a comprehensive, population-based case-control investigation of underlying causes for autism and triggers of regression. Cases are children with autism, and two other groups are included: children with developmental delay and children selected at random from the general population. Close to 700 participants will have been enrolled in CHARGE by the end of the first funding period of the CCEH. The proposed continuation of the CHARGE study will use the R01 mechanism due to the high cost of recruitment and data collection, which is too expensive to maintain within the CCEH, and will extend the sample size to 1600. The larger sample size will permit the examination of exposures of relatively low prevalence, of gene- environment interactions, and of etiologic specificity for subtypes of autism defined by phenotypic characteristics. The proposed study will address the possible role of exposures during the peri-conceptional, gestational, perinatal, and early childhood periods, including infections, assisted reproductive technology, medications, pesticides, brominated flame retardants, and metals. It will also evaluate these environmental factors in conjunction with genes involved in xenobiotic metabolism or suggested by our current microarray results or by SNP arrays. Finally, because autism manifests in heterogeneous phenotypes, we will examine all exposures in relation to specific subtypes of autism defined by developmental patterns such as regression versus early onset, high versus. low adaptive/cognitive function, and by immunologic profiles. This project addresses a critical and growing public health concern, namely, autism; this disorder affects as many as one in 166 children. Our project will evaluate risk factors that may play a causal role.